Any experience from new release of hive, i found previous one unstable so im still on 0.6-203@210424
There seems to be many fixes for TRM. Have you noticed any benefits from upgrading??
I’m actually finding 0.6-203@210426 to be somewhat unstable…
TRM seems to be bugging out intermittently and I have to reboot the rig.
Hiveos was rock solid for me before this upgrade.
1 Like
i went to 0.6-202@210426 now no errors, stable long run after downgrade. Started to get dead gpu errors on 203 so downgraded…
Downgrading as I’m writing this, hopefully there’ll be a fix out soon.
1 Like
Still getting crashes of TRM at 1 hour intervals. Can’t recover by restarting miner but I have to reboot entire rig.
I wonder is trm aware of these issues…i allso must go way down from normal oc to avoid dead gpu error… this is not good. someone said that lolminer is working fine instead trm for now, i have cards that hash really bad with lolminer so i must stay with trm… not happy with trm now
I’ve managed somehow to keep it from crashing for over 5,5 hours. Don’t want to jinx it but have a good feeling. Running TRM 0.20.1 as of now.
1 Like
nice, i havent had too many issues either, some anyway so i added to trm flightsheet-
–init_style=3
–watchdog_disabled
1 Like
thx I’ll look into it, Trex github seem sto be down for me and I need to get this fixed 
managed to bood up with init 3 fine few times but looks like if oc ´s are too high it wont like that parameter…still testing… edit. yes…it seem that if my first gpu oc is “normal” trm will hang on that one and the whole boot is aborted. init 3 will initialiýze all gpus att the same time, but still no go if i use my default oc on gpu 1…so deleted init config and go without trm watchdog and use only hive hash/algo wathcdog…maybe that will eliminate dead gpu and other "unnessesary issues with trm…
init config seems to be working fine with the rig with only amd 5xx gpus,now time will tell if dead gpus are gone and i can use my normal oc´s, that are not super high…quite conservative.Would be nice to get from 580 and 590 gpus around 32 mh stable
I’m experiencing the same but with 3060Ti, 3070 and two 1070’s, I don’t have that crazy overclocks. I remember there was some software out there that dropped overclocks when rebuilding DAG or whatever.
from trm manual
could it be something like: –eth_stagger Enables staggering of gpus when building a new DAG. This is more lean on PSUs that don’t like going from 0-100% load on all gpus at the same time. ??
well…now i see, thats more psu related… interesting options anyways
Seems like I had memory strap 2 for my gtx 1070’s thought I had removed that. Will check back if setting it to zero solves the issue.
1 Like
Have you managed to maintain stability?
with conservative clocks and now it seems that disabling trm watchdog too away last errors. are u sure that u have enought big psu and if your mb requires, is there power in pci lanes. Canyt help too much cause i dont have experience of nvidia
I have plenty of PSU room, running on B250 mining expert with 10 cards.
1 Like
i had my both rigs un stabile for nearly 2 months man…24/7 in the basement tuning and trying to make them work…somehow i just manage to get them stabile…now they are just making money…i dont know what to do with my time 
…so i try to help others with similar issues 
Seems to be stable for now, not a crash for 10h which is somewhat of record as of lately.
with TRM? what was your issues again before, just random crashes? are u on latest hive OS? .and latest TRM? I went to TRM 0.8.0… i really dont know whats wrong…i got 48hrs stable runs on both rigs and boom…all these issues, changed 3 risers today,even im sure that there is nothing wron with them. well,it is what it is ! some day buddy…some day 
I updated to the latest hiveos but donwgraded my TRM version to 0.8.2. It seems more stable for me tha the one supplied in the latest hiveos update
1 Like